Benefits of using basic, imprecise or uncertain data for elaborating sewer inspection programmes

Abstract

One key goal of sewer inspection programmes is to target segments in the worst condition. Despite the development of deterioration models, the influence of available data on models’ predictive power has not been studied in depth yet. In this article, numerical experiments have been conducted to answer three main questions: (1) How can the data most probably available within a utility be used to define an effective inspection programme? (2) Can we use an auxiliary variable in order to compensate effects of missing data on inspection programmes? (3) Is it worth to accept a degree of uncertainty within data instead of not having them? In other words, is it preferable to have uncertainty instead of incompleteness within utility database? In order to respond to these questions, we considered an asset stock and then degraded the information by introducing uncertainty, imprecision and incompleteness within, to form a utility's database. The results show that significant improvement of inspection programmes could be achieved by using the most probably available data within utilities. We also show that using the notion of ‘district’ can provide efficient results when the most informative factor ‘age’ is not available. Finally, it is shown that having uncertain data is preferable to having incompleteness.

Keywords::

• assets
• decision support system
• data analysis
• regression analysis
• failures
• sewers and drains
• inspection

Notes

^1. Email: [email protected]

^2. Email: [email protected]

^3. Email: [email protected]